Alterations in power spectral density in motor- and pain-related networks on neuropathic pain after spinal cord injury

Abstract

Background: The mechanisms by which mobility function and neuropathic pain are mutually influenced by supraspinal plasticity in motor- and pain-related brain networks following spinal cord injury (SCI) remains poorly understood.

Objective: To determine cortical and subcortical resting-state network alterations using power spectral density (PSD) analysis and investigate the relationships between these intrinsic alterations and mobility function and neuropathic pain following SCI.

Methods: A total of 41 patients with incomplete SCI and 33 healthy controls were included. The degree of mobility and balance function and severity of neuropathic pain and depressive mood were evaluated. The resting-state functional magnetic resonance imaging data of low-frequency fluctuations were analyzed based on PSD. Differences in PSD values between patients with SCI and controls were assessed using the two-sample t-test (false discovery rate-corrected P < 0.05). The relationship between PSD values and mobility function and pain intensity was assessed using Pearson's correlation coefficient adjusted for the severity of depressive mood.

Results: Compared with healthy controls, lower PSD values in supplementary motor and medial prefrontal areas (the anterior cingulate cortex, ventral medial prefrontal cortex, and superior orbito-prefrontal cortex) were associated with greater pain severity and poorer postural balance and mobility (P < 0.05) in patients with SCI, whereas higher PSD values in the primary motor cortex, premotor cortex, thalamus, and periaqueductal gray were associated with greater pain severity and poorer postural balance and mobility (P < 0.05).

Conclusions: Cortical and subcortical plastic alterations in intrinsic motor- and pain-related networks were observed in patients with SCI and were simultaneously associated with neuropathic pain intensity and degree of mobility function.

Keywords: Mobility; Neuropathic pain; Power spectral density; Resting-state functional magnetic resonance imaging; Spinal cord injury.

Fig. 1

Power spectral density (PSD) map of a low-frequency band (0.01–0.1 Hz) in the resting brain of patients with spinal cord injury (SCI) (right) and healthy controls (left). The map was thresholded at an FDR-corrected P-value of <0.05 with a minimum cluster size of 10.

Fig. 2

Group differences in the power spectral density (PSD) of low-frequency fluctuations (0.01–0.1 Hz). The top map shows lower PSD values in patients with incomplete SCI than that in healthy controls. The bottom map shows higher PSD values in patients with SCI than that in healthy controls. The map was thresholded at an FDR-corrected P-value of <0.05 with a minimum cluster size of 10.

Fig. 3

Correlation between the mean Z-score of power spectral density (PSD), showing lower PSD values in patients with SCI than that in controls, degree of motor function, and neuropathic pain. The mean Z-score of PSD in the left superior orbito-prefrontal cortex (A), anterior cingulate cortex (B), right ventral medial prefrontal cortex (C), left medial prefrontal cortex (D), right middle frontal gyrus (E), and left supplementary motor area (F) were correlated with VAS, BBS, and/or FAC scores (*P < 0.05, ^**P < 0.001).

Fig. 4

Correlation between the mean Z-score of power spectral density (PSD), showing higher values of PSD in patients with SCI than that in controls, degree of motor function, and neuropathic pain. The mean Z-score of PSD in the left midbrain (A), left thalamus (B), right precentral gyrus (C), and left premotor cortex (D) were correlated with VAS, BBS, and/or FAC scores (*P < 0.05, ^**P < 0.001).